Harden VedAstro adapter and index PyJHora artifacts

This commit is contained in:
732642856
2026-06-27 14:26:28 +08:00
parent 6f46805fab
commit eb05dc3959
6 changed files with 433 additions and 9 deletions
@@ -0,0 +1,131 @@
# Antigravity AI Sidecar Work Order - Round 39 - 2026-06-27
## Mission
本轮继续只服务印度占星 `skill` 主线,不做网页/app 外观工作,不改核心源码。目标是替主线程进一步压缩“全球第一 skill”还缺的硬情报,尤其围绕:
1. PyJHora black-box 资产继续扩充
2. VedAstro service adapter 对接前的字段映射与公开契约核查
3. Dasha / Shadbala / Tajika 三大战线的真实 external oracle 首包之后的第二批优先级
4. 文章级解释模板的权威降噪
5. 长期 benchmark 护城河的公开字段设计
## Hard Constraints
- 不修改主仓核心实现代码。
- 只做研究、检索、黑盒比对、任务拆解、资料整理。
- 必须优先复用现有 `docs/research/``references/oracle/``references/open_source_sources/` 资料。
- AGPL / GPL 只允许 black-box、stdout、字段映射、行为观察,不允许建议复制代码实现。
- 每份报告必须能直接减少 Codex 下一轮的上下文浪费。
## Work Packages
### A. PyJHora black-box asset expansion board
目标:
- 基于现有 `references/oracle/artifacts/``pending_packets/`,列出下一批最值得补抓的 12 个 PyJHora black-box 资产。
- 优先级偏向:
- Chara Dasha
- Kalachakra Dasha
- Ashtakavarga / SAV / BAV
- Bhava Bala
- Special Lagnas
- Vargottama / Pushkara / Avastha 边界例子
输出:
- `antigravity_round39_pyjhora_blackbox_asset_expansion_board_2026_06_27.md`
### B. VedAstro request/response mapping audit
目标:
- 只研究 VedAstro 作为 service adapter candidate 时,哪些字段和我们本地 contract 已对齐,哪些还缺中间映射层。
- 明确:
- ayanamsa policy
- node policy
- body naming
- divisional chart identifiers
- dasha naming
- timezone / DST expectations
输出:
- `antigravity_round39_vedastro_mapping_audit_2026_06_27.md`
### C. Three-front second-wave packet rerank
目标:
- 在 Dasha / Shadbala / Tajika 三大战线里,不再讨论 first packet,而是排出 second-wave 的前 15 个 packet。
- 每个 packet 说明:
- 为什么它是 second-wave 高 ROI
- 需要人类录入还是可以先靠公开书例
- 对 skill 精度和“全球第一”目标的实际贡献
输出:
- `antigravity_round39_three_front_second_wave_packets_2026_06_27.md`
### D. Authority-first article template truth cleanup
目标:
- 把近几轮用户补充的文章、截图线索里,容易误导的说法筛一遍。
- 特别检查:
- 财富点 / Lakshmi / Dhana
- DK / UL / spouse elevation
- 上升度数 / 紧密合相
- Nakshatra 天赋标签
- 工作时间预测
输出:
- `antigravity_round39_article_truth_cleanup_2026_06_27.md`
### E. Public benchmark schema v2 board
目标:
- 设计一个别人难以短期追上的公开 benchmark 字段体系。
- 不是产品方案,而是字段和数据制度方案:
- case metadata
- external evidence provenance
- tolerance policy
- ayanamsa / node / timezone normalization
- explanation-template scoring
输出:
- `antigravity_round39_public_benchmark_schema_v2_2026_06_27.md`
### F. Codex direct top 80
目标:
- 基于以上 5 包,为主线程输出只服务 skill 的 Top 80 下一步动作。
- 每条动作必须标记:
- `doc_freeze`
- `tdd_now`
- `needs_human_oracle`
- `blocked_by_license`
输出:
- `antigravity_round39_codex_round40_skill_top80_2026_06_27.md`
## Deliverables
至少生成以下 6 个文件到 `docs/research/`
1. `antigravity_round39_pyjhora_blackbox_asset_expansion_board_2026_06_27.md`
2. `antigravity_round39_vedastro_mapping_audit_2026_06_27.md`
3. `antigravity_round39_three_front_second_wave_packets_2026_06_27.md`
4. `antigravity_round39_article_truth_cleanup_2026_06_27.md`
5. `antigravity_round39_public_benchmark_schema_v2_2026_06_27.md`
6. `antigravity_round39_codex_round40_skill_top80_2026_06_27.md`
## Success Standard
- 不重复“还有很多 gap”这种空话。
- 每份报告都能直接节省 Codex 下一轮时间。
- 明确区分:
- 现在可直接编码
- 现在只能文档冻结
- 必须等待人类 external oracle
- 受许可证隔离,只能 black-box
@@ -0,0 +1,76 @@
{
"scope": "pyjhora_oracle_artifact_manifest",
"generated_at": "2026-06-27T06:25:10.958318+00:00",
"artifact_count": 9,
"packet_count": 8,
"artifacts": [
"pyjhora_extreme_latitude_kp_shadbala_stdout_20260627.txt",
"pyjhora_redacted_place_shadbala_raman_stdout_20260627.txt",
"pyjhora_historical_epoch_dasha_stdout_20260627.txt",
"pyjhora_oracle_artifact_manifest.json",
"pyjhora_steve_jobs_dasha_stdout_20260627.txt",
"pyjhora_steve_jobs_shadbala_lahiri_stdout_20260627.txt",
"pyjhora_steve_jobs_varshaphala_1984_lahiri_stdout_20260627.txt",
"pyjhora_user_REDACTED_YEAR_dasha_stdout_20260627.txt",
"pyjhora_user_REDACTED_YEAR_shadbala_lahiri_stdout_20260627.txt"
],
"pending_packets": [
"external_template_extreme_latitude_kp_pyjhora_20260627.json",
"external_template_redacted_place_shadbala_raman_pyjhora_20260627.json",
"external_template_historical_epoch_lahiri_pyjhora_20260627.json",
"external_template_steve_jobs_dasha_lahiri_pyjhora_20260627.json",
"external_template_steve_jobs_shadbala_lahiri_pyjhora_20260627.json",
"external_template_steve_jobs_varshaphala_1984_lahiri_pyjhora_20260627.json",
"external_template_user_REDACTED_YEAR_moon_longitude_lahiri_pyjhora_20260627.json",
"external_template_user_REDACTED_YEAR_shadbala_lahiri_pyjhora_20260627.json"
],
"fronts": {
"dasha": {
"artifact_count": 4,
"packet_count": 3,
"artifacts": [
"pyjhora_historical_epoch_dasha_stdout_20260627.txt",
"pyjhora_oracle_artifact_manifest.json",
"pyjhora_steve_jobs_dasha_stdout_20260627.txt",
"pyjhora_user_REDACTED_YEAR_dasha_stdout_20260627.txt"
],
"pending_packets": [
"external_template_extreme_latitude_kp_pyjhora_20260627.json",
"external_template_historical_epoch_lahiri_pyjhora_20260627.json",
"external_template_steve_jobs_dasha_lahiri_pyjhora_20260627.json"
]
},
"shadbala": {
"artifact_count": 4,
"packet_count": 4,
"artifacts": [
"pyjhora_extreme_latitude_kp_shadbala_stdout_20260627.txt",
"pyjhora_redacted_place_shadbala_raman_stdout_20260627.txt",
"pyjhora_steve_jobs_shadbala_lahiri_stdout_20260627.txt",
"pyjhora_user_REDACTED_YEAR_shadbala_lahiri_stdout_20260627.txt"
],
"pending_packets": [
"external_template_redacted_place_shadbala_raman_pyjhora_20260627.json",
"external_template_steve_jobs_shadbala_lahiri_pyjhora_20260627.json",
"external_template_user_REDACTED_YEAR_moon_longitude_lahiri_pyjhora_20260627.json",
"external_template_user_REDACTED_YEAR_shadbala_lahiri_pyjhora_20260627.json"
]
},
"tajika_sahams": {
"artifact_count": 1,
"packet_count": 1,
"artifacts": [
"pyjhora_steve_jobs_varshaphala_1984_lahiri_stdout_20260627.txt"
],
"pending_packets": [
"external_template_steve_jobs_varshaphala_1984_lahiri_pyjhora_20260627.json"
]
}
},
"files": {
"manifest": "references/oracle/artifacts/pyjhora_oracle_artifact_manifest.json",
"artifacts_dir": "references/oracle/artifacts",
"pending_packets_dir": "references/oracle/artifacts/pending_packets"
},
"boundary": "These files are black-box external evidence only. They document PyJHora outputs and pending oracle packets without importing AGPL code into the local skill implementation."
}
@@ -0,0 +1,88 @@
#!/usr/bin/env python3
"""Generate a tracked manifest for PyJHora black-box oracle artifacts."""
from __future__ import annotations
import json
from collections import defaultdict
from datetime import datetime, timezone
from pathlib import Path
from typing import Any
ROOT = Path(__file__).resolve().parents[1]
ARTIFACTS_DIR = ROOT / "references" / "oracle" / "artifacts"
PENDING_DIR = ARTIFACTS_DIR / "pending_packets"
OUTPUT_PATH = ARTIFACTS_DIR / "pyjhora_oracle_artifact_manifest.json"
def _front_for_name(name: str) -> str:
lowered = name.lower()
if "varshaphala" in lowered or "sahams" in lowered or "tajika" in lowered:
return "tajika_sahams"
if "shadbala" in lowered or "moon_longitude" in lowered:
return "shadbala"
return "dasha"
def _relative(path: Path) -> str:
return str(path.relative_to(ROOT))
def build_manifest() -> dict[str, Any]:
artifacts = sorted(
path for path in ARTIFACTS_DIR.glob("pyjhora_*")
if path.is_file()
)
pending_packets = sorted(
path for path in PENDING_DIR.glob("*pyjhora_20260627.json")
if path.is_file()
)
fronts: dict[str, dict[str, Any]] = defaultdict(lambda: {"artifact_count": 0, "packet_count": 0, "artifacts": [], "pending_packets": []})
for artifact in artifacts:
front = _front_for_name(artifact.name)
fronts[front]["artifact_count"] += 1
fronts[front]["artifacts"].append(artifact.name)
for packet in pending_packets:
front = _front_for_name(packet.name)
fronts[front]["packet_count"] += 1
fronts[front]["pending_packets"].append(packet.name)
report = {
"scope": "pyjhora_oracle_artifact_manifest",
"generated_at": datetime.now(timezone.utc).isoformat(),
"artifact_count": len(artifacts),
"packet_count": len(pending_packets),
"artifacts": [path.name for path in artifacts],
"pending_packets": [path.name for path in pending_packets],
"fronts": {
front: {
"artifact_count": payload["artifact_count"],
"packet_count": payload["packet_count"],
"artifacts": payload["artifacts"],
"pending_packets": payload["pending_packets"],
}
for front, payload in sorted(fronts.items())
},
"files": {
"manifest": _relative(OUTPUT_PATH),
"artifacts_dir": _relative(ARTIFACTS_DIR),
"pending_packets_dir": _relative(PENDING_DIR),
},
"boundary": (
"These files are black-box external evidence only. They document PyJHora outputs and "
"pending oracle packets without importing AGPL code into the local skill implementation."
),
}
OUTPUT_PATH.write_text(json.dumps(report, ensure_ascii=False, indent=2) + "\n", encoding="utf-8")
return report
def main() -> int:
print(json.dumps(build_manifest(), ensure_ascii=False, indent=2))
return 0
if __name__ == "__main__":
raise SystemExit(main())
+21 -9
View File
@@ -11,6 +11,7 @@ from __future__ import annotations
import argparse
import json
import os
import socket
from pathlib import Path
from typing import Any
from urllib import request, error
@@ -59,6 +60,7 @@ PARITY_CASES = {
}
DEFAULT_TIMEOUT_SECONDS = 8
TIMEOUT_ENV = "VEDASTRO_TIMEOUT_SECONDS"
RETRY_POLICY = {
"max_attempts": 2,
"backoff_seconds": 1,
@@ -67,6 +69,16 @@ RETRY_POLICY = {
ALLOW_NETWORK_ENV = "VEDASTRO_ENABLE_NETWORK"
def _timeout_seconds() -> float:
raw = os.environ.get(TIMEOUT_ENV, "").strip()
if not raw:
return DEFAULT_TIMEOUT_SECONDS
try:
return float(raw)
except ValueError:
return DEFAULT_TIMEOUT_SECONDS
def schema() -> dict[str, Any]:
request_example = {
**PARITY_CASES["beijing_first_use_demo"],
@@ -130,7 +142,7 @@ def _unconfigured(reason: str) -> dict[str, Any]:
"endpoint_env": "VEDASTRO_API_ENDPOINT",
"api_key_env": "VEDASTRO_API_KEY",
"provenance_mode": "external_service_candidate",
"timeout_seconds": DEFAULT_TIMEOUT_SECONDS,
"timeout_seconds": _timeout_seconds(),
"retry_policy": RETRY_POLICY,
},
}
@@ -156,7 +168,7 @@ def _normalize_success(payload: dict[str, Any], endpoint: str) -> dict[str, Any]
"transport": "http_json_service_boundary",
"endpoint": endpoint,
"provenance_mode": "external_service_candidate",
"timeout_seconds": DEFAULT_TIMEOUT_SECONDS,
"timeout_seconds": _timeout_seconds(),
"retry_policy": RETRY_POLICY,
**(payload.get("source_metadata") or {}),
},
@@ -189,7 +201,7 @@ def run_case(case_id: str) -> dict[str, Any]:
"transport": "http_json_service_boundary",
"endpoint": endpoint,
"provenance_mode": "external_service_candidate",
"timeout_seconds": DEFAULT_TIMEOUT_SECONDS,
"timeout_seconds": _timeout_seconds(),
"retry_policy": RETRY_POLICY,
"network_execution_env": ALLOW_NETWORK_ENV,
},
@@ -201,7 +213,7 @@ def run_case(case_id: str) -> dict[str, Any]:
headers={"Content-Type": "application/json"},
method="POST",
)
with request.urlopen(req, timeout=DEFAULT_TIMEOUT_SECONDS) as resp:
with request.urlopen(req, timeout=_timeout_seconds()) as resp:
raw = resp.read().decode("utf-8")
payload = json.loads(raw)
except error.HTTPError as exc:
@@ -215,7 +227,7 @@ def run_case(case_id: str) -> dict[str, Any]:
"transport": "http_json_service_boundary",
"endpoint": endpoint,
"provenance_mode": "external_service_candidate",
"timeout_seconds": DEFAULT_TIMEOUT_SECONDS,
"timeout_seconds": _timeout_seconds(),
"retry_policy": RETRY_POLICY,
"network_execution_env": ALLOW_NETWORK_ENV,
},
@@ -231,12 +243,12 @@ def run_case(case_id: str) -> dict[str, Any]:
"transport": "http_json_service_boundary",
"endpoint": endpoint,
"provenance_mode": "external_service_candidate",
"timeout_seconds": DEFAULT_TIMEOUT_SECONDS,
"timeout_seconds": _timeout_seconds(),
"retry_policy": RETRY_POLICY,
"network_execution_env": ALLOW_NETWORK_ENV,
},
}
except TimeoutError:
except (TimeoutError, socket.timeout):
return {
"backend": "vedastro_service_adapter_candidate",
"available": False,
@@ -247,7 +259,7 @@ def run_case(case_id: str) -> dict[str, Any]:
"transport": "http_json_service_boundary",
"endpoint": endpoint,
"provenance_mode": "external_service_candidate",
"timeout_seconds": DEFAULT_TIMEOUT_SECONDS,
"timeout_seconds": _timeout_seconds(),
"retry_policy": RETRY_POLICY,
"network_execution_env": ALLOW_NETWORK_ENV,
},
@@ -263,7 +275,7 @@ def run_case(case_id: str) -> dict[str, Any]:
"transport": "http_json_service_boundary",
"endpoint": endpoint,
"provenance_mode": "external_service_candidate",
"timeout_seconds": DEFAULT_TIMEOUT_SECONDS,
"timeout_seconds": _timeout_seconds(),
"retry_policy": RETRY_POLICY,
"network_execution_env": ALLOW_NETWORK_ENV,
},
@@ -0,0 +1,37 @@
from __future__ import annotations
import json
import subprocess
import sys
from pathlib import Path
ROOT = Path(__file__).resolve().parents[1]
def test_pyjhora_oracle_artifact_manifest_reports_current_blackbox_assets() -> None:
completed = subprocess.run(
[sys.executable, "scripts/generate_pyjhora_oracle_artifact_manifest.py"],
cwd=ROOT,
text=True,
capture_output=True,
timeout=120,
check=False,
)
assert completed.returncode == 0, completed.stderr or completed.stdout
report = json.loads(completed.stdout)
assert report["scope"] == "pyjhora_oracle_artifact_manifest"
assert report["artifact_count"] >= 5
assert report["packet_count"] >= 4
assert report["fronts"]["dasha"]["artifact_count"] >= 2
assert report["fronts"]["shadbala"]["artifact_count"] >= 2
assert report["fronts"]["tajika_sahams"]["artifact_count"] >= 1
assert report["files"]["manifest"].endswith("references/oracle/artifacts/pyjhora_oracle_artifact_manifest.json")
manifest = ROOT / "references" / "oracle" / "artifacts" / "pyjhora_oracle_artifact_manifest.json"
assert manifest.exists()
manifest_data = json.loads(manifest.read_text(encoding="utf-8"))
assert manifest_data["scope"] == "pyjhora_oracle_artifact_manifest"
assert "pyjhora_steve_jobs_dasha_stdout_20260627.txt" in manifest_data["artifacts"]
assert "external_template_steve_jobs_dasha_lahiri_pyjhora_20260627.json" in manifest_data["pending_packets"]
@@ -198,3 +198,83 @@ def test_vedastro_service_adapter_classifies_http_error() -> None:
report = json.loads(completed.stdout)
assert report["status"] == "http_error"
assert "503" in report["reason"]
def test_vedastro_service_adapter_classifies_invalid_json_response() -> None:
class Handler(BaseHTTPRequestHandler):
def do_POST(self) -> None: # noqa: N802
body = b"not-json"
self.send_response(200)
self.send_header("Content-Type", "application/json")
self.send_header("Content-Length", str(len(body)))
self.end_headers()
self.wfile.write(body)
def log_message(self, format: str, *args) -> None: # noqa: A003
return
server = HTTPServer(("127.0.0.1", 0), Handler)
thread = threading.Thread(target=server.serve_forever, daemon=True)
thread.start()
try:
env = os.environ.copy()
env["VEDASTRO_API_ENDPOINT"] = f"http://127.0.0.1:{server.server_port}/vedastro"
env["VEDASTRO_ENABLE_NETWORK"] = "1"
completed = subprocess.run(
[sys.executable, "scripts/vedastro_service_adapter.py", "--case", "beijing_first_use_demo"],
cwd=ROOT,
text=True,
capture_output=True,
timeout=120,
check=False,
env=env,
)
finally:
server.shutdown()
thread.join(timeout=5)
assert completed.returncode == 0, completed.stderr or completed.stdout
report = json.loads(completed.stdout)
assert report["status"] == "invalid_json"
def test_vedastro_service_adapter_classifies_timeout() -> None:
class Handler(BaseHTTPRequestHandler):
def do_POST(self) -> None: # noqa: N802
import time
time.sleep(0.2)
body = b"{}"
self.send_response(200)
self.send_header("Content-Type", "application/json")
self.send_header("Content-Length", str(len(body)))
self.end_headers()
self.wfile.write(body)
def log_message(self, format: str, *args) -> None: # noqa: A003
return
server = HTTPServer(("127.0.0.1", 0), Handler)
thread = threading.Thread(target=server.serve_forever, daemon=True)
thread.start()
try:
env = os.environ.copy()
env["VEDASTRO_API_ENDPOINT"] = f"http://127.0.0.1:{server.server_port}/vedastro"
env["VEDASTRO_ENABLE_NETWORK"] = "1"
env["VEDASTRO_TIMEOUT_SECONDS"] = "0.05"
completed = subprocess.run(
[sys.executable, "scripts/vedastro_service_adapter.py", "--case", "beijing_first_use_demo"],
cwd=ROOT,
text=True,
capture_output=True,
timeout=120,
check=False,
env=env,
)
finally:
server.shutdown()
thread.join(timeout=5)
assert completed.returncode == 0, completed.stderr or completed.stdout
report = json.loads(completed.stdout)
assert report["status"] == "timeout"